Esotropiart

Has anyone besides me noticed that Windows XP sorts files completely different than any previous version of Windows? In fact, this complete change seems to have gone by mostly unnoticed. Since I deal with files and folders numbering in the thousands every day, I immediately noticed the first time I used My Computer. To be honest, this new sorting method bugs me. By the way, if you are not interested in such things or think they are negligible, then skip reading this. It will bore you out of your mind! For people like me who work with computers and numbers day-in, day-out, it is very significant and pertinent.

Previously, Windows, DOS and virtually every application developed for a PC environment sorted files using a simple case-insensitive ASCII sort. This means that the order of the files is determined by comparing the characters of the names to the order the characters are found in an ASCII character chart, a universally standard order (yes, it's an old standard, but the order still stands). For example, numbers 0 to 9 fall before letters A to Z in the chart, and therefore, they will come first in a list. Symbols have their own special place in the chart: some come before the numbers and letters, some fall after. Since symbols are not regularly used at the beginning and end of filenames, it's not overly important to memorize the entire ASCII character chart. It is simply important to know that numbers come before letters. When I learned to do office filing many years ago, such considerations were followed, and even the phone book puts numbers before letters. It seems to me to be a sensible and good standard to adhere to even through thousands of eons when computers have evolved into intelligent red gel that one can paint on any surface to make wishes come true. Why change it? Naturally some people have decided to consider numbers to be sorted with the first letter of the number as it would be written out. For instance, one would sort with the Os, Two and Three with the Ts, Four with the Fs, etc. Many old dictionaries (and some telephone books) would sort 24 with T, and actually commit the atrocity of spelling it out, twenty-four. This is an atrocity because it actually changes the term or business name. I work at C2F, not CTwoF, or CeeTwoEf, or SeaTooEffe, etc.

Here is an example of a case-insensitive ASCII sort:
(the way things have always been done in a PC environment file sort - should look familiar to most computer users)

1. ART0301XX.JPG
2. ART0401ST.JPG
3. ART040403.JPG
4. ART040410.JPG
5. ART0442.JPG
6. art0475.jpg
7. ART1202.JPG
8. ART124008.JPG
9. ART1675.JPG
10. art3012.jpg
11. ART310027.JPG
12. ART310048.JPG
13. ART5311D.JPG
14. art6625cr.jpg
15. ART80121.JPG
16. ART828.JPG
17. ART869.JPG
18. ART8732C.JPG
19. ARTB906.JPG
20. ARTE215.JPG
21. ARTE277.JPG
22. ARTS242.JPG
23. ARTS8740.JPG
24. ARTS87401250.JPG
25. ARTS8787

Here is the same list of filenames as sorted by Windows XP:

1. ART0301XX.JPG
2. ART0401ST.JPG
3. ART0442.JPG
4. art0475.jpg
5. ART828.JPG
6. ART869.JPG
7. ART1202.JPG
8. ART1675.JPG
9. art3012.jpg
10. ART5311D.JPG
11. art6625cr.jpg
12. ART8732C.JPG
13. ART040403.JPG
14. ART040410.JPG
15. ART80121.JPG
16. ART124008.JPG
17. ART310027.JPG
18. ART310048.JPG
19. ARTB906.JPG
20. ARTE215.JPG
21. ARTE277.JPG
22. ARTS242.JPG
23. ARTS8740.JPG
24. ARTS8787
25. ARTS87401250.JPG

Basically, the new XP sort is attempting to look at filenames as numerically as possible. I'm sure it was changed on a whim as a "solution" to the ASCII sort's inability to sort series of pictures based on a number suffix. Since 10 and 15 actually come before 2 in an ASCII sort, in order to make files sort in order of their numeric value, a common number of digits must be used in a series, filling the extra empty digits with zeros. See here...

An ASCII sort of a series of a series of images with common convention, but no leading zeroes on the numbers:

1. image1.jpg
2. image10.jpg
3. image11.jpg
4. image15.jpg
5. image2.jpg
6. image20.jpg
7. image3.jpg
8. image4.jpg

A Windows XP sort on the same list of images:

1. image1.jpg
2. image2.jpg
3. image3.jpg
4. image4.jpg
5. image10.jpg
6. image11.jpg
7. image15.jpg
8. image20.jpg

Now, at first glance at this second sort, most people would think, "Ah, fantastic! The solution to my woes! Now I don't have to put leading zeroes before image series numbers to make them sort right." Here is an example of leading zeroes to make series sort as most users would wish in an ASCII sorting environment:

1. image01.jpg
2. image02.jpg
3. image03.jpg
4. image04.jpg
5. image10.jpg
6. image11.jpg
7. image15.jpg
8. image20.jpg

Here's my personal opinion on the matter. No, it's not hard fact, but opinion and preference. I prefer the ASCII sort because it is far more predictable. A human can always find an image in an ASCII sorted list because, based on the fact that 0 to 9 always come before A to Z, it is easy to use one's knowledge of the alphabet to find any filename imaginable. On the other hand, the new XP sort might occasionally make it easier to visualize a large series of similar filenames as seen above. However, the XP sort only benefits a sort when every single image has a common naming convention. In other words, all images must have the same first characters before the numbers, or the whole numeric sort will break down and be absolutely meaningless. In fact, if the filenames aren't named exactly the same with a common series, the XP sort will appear to almost jumble the filenames in a completely sporadic and impossible to discern manner. Take the example of the second list above, showing the XP sort on the images starting with "ART...". That is actually how the files will be sorted in Windows XP. Notice that XP assumes that all numbers contained inside a filename are a series, so it places the lowest value number at the front, even if there are different characters and numbers that follow. In the case of these files, they are not in a series at all. In fact, the numeric part of the filename is a product number, not a series. Each number has virtually nothing to do with the others. So the sort is not sensible or helpful for this application.

Try looking at a folder full of thousands of images containing different number combinations in their names, and you will get dizzy trying to find anything at all. I understand exactly how the XP sort works, have a detail-oriented mind, and have worked with computers for over 15 years, and have a hard time finding files in a large folder in Windows XP. It has nothing to do with teaching an old dog new tricks because I typically have no problem picking up new stuff. My mind isn't that old yet :). This new sort is only to fix that little problem of commonly named series so that it is no longer necessary to add leading zeroes. This leading zero issue isn't really much of an issue because most people that work with computers for any time at all become very aware that they have to use leading zeroes to make their series order properly. Besides, it looks nicer and more consistent for filenames to have the same number of digits so their lengths line up horizontally anyway. And this new sort method messes up every other possible conception of order and sense that the human mind could possibly conceive.

Therefore, if this numeric, anti-intelligent sort method is to be used at all, there should at least be a checkbox option somewhere in Windows XP browsing options to allow people to disable it and return to the more sensible and easy (in my opinion) and consistent (without a doubt, absolute fact) ASCII sort. Instead, the only way to return to the old sorting method is to hack into the registry. For those who care about this and would like to revert to the ASCII sorting method, click here to visit Microsoft's documented solution. However, don't bother with this if you are not a computer geek. Editing the Windows Registry is very risky business - not for the novice. Changing the wrong registry key can cause irreparable damage to your computer, so avoid it unless you know exactly what you are doing.

Note: In my humble opinion, the purpose of sorting is to translate computer information into a format that makes it easier and faster for human eyes and minds to locate and utilize. If a sort method only confuses the issue and makes it take 10 minutes to find a file due to a confusing method where it would take 30 seconds using a standardized method, why use such a method? When browsing through a folder with thousands of images named as shown above, it is actually faster to use a search, rather than knowing where something falls in the list with the XP sort, so the sort is actually worthless. Ah well. I'm sure there are many out there who would disagree. Just look at the second list on this page and try to make sense of it (and imagine it is 6000 members long!). I understand both the ASCII and XP sorts, but knowing the XP sort doesn't make it any easier to find a given filename in a list. You have to know in advance the number of digits in a filename's number, and all the surrounding characters. In short, it takes a lot of thought. Hmmmm... would I rather use one simple rule that I learned in kindergarden "ABCDEFG, HIJK, LMNOP...", or sit and think of what number might be contained in a filename and where? They may as well be randomized! I apologize again to those who don't care or are sickened by technical babble :) God bless and happy sorting!